This revised submission of the Clinical Trials Section of the Silvio O. Conte Neuroscience Center for the study of Glutamate Dysregulation in Schizophrenia has been enhanced by a new collaboration with Dr. Daniel Javitt. We will extend findings produced during the first five years and will examine possible explanations for the negative findings of our six month trial with D-cycloserine and the "CONSIST" trial, a 4 month comparison of glycine and D-cycloserine-- neither found effects on negative symptoms or cognition. Because neither D-cycloserine nor glycine may adequing Dr. Javitt's IND, we will conduct a placebo-controlled trial of D-serine a glycine site partial agonist, did not produce siginificant improvement in negative or cognitive symptoms will be followed-up by a new approach in which D-cycloserine is administered once-weekly. This design is intended to avoid tolerance to repeated dosing with D-cycloserine; animal studies and recent studies in patients with phobia indicate that single doses substantially improve cognitive function, but that tolerance develops after approximately two weeks of daily dosing. We will conduct a trial of D-serine co-treatment. This study follows from a previous finding by Tsai and colleagues that the more potent, full agonist D-serine may be effective for positive, negative and cognitive symptoms without evidence for tolerance with repeated dosing. The recent completion of toxicology studies will allow us to obtain an IND for study of D-serine within the next year. In collaboration with Dr. Yurgelun- Todd, we will perform fMRI at baseline and week 8 of the D-serine trial to extend our previous finding of enhanced temporal lobe activation with D-cycloserine which correlated with improvement of negative symptoms. We will expand our collection of DNA samples and our extensive database of phenotypic information from 200 to 500 patients to allow examination of alleles relevant to glutamatergic regulation in relation to clinical characteristics and response to glutamatergic agents. Study of polymorphisms of G72 and D-amino acid oxidase as predictors of D-serine response is one example. We will also further pursue our studies of GCPII by examining two additional polymorphisms in a larger sample. In collaboration with Dr. Yurgelun-Todd, we will also examine NAA/NAAG concentrations in prefrental cortex as a potential biological marker of GCPII activity.